Show HN: I made a computer vision tool for evaluating deadlift form (github.com via hn) gemma 6w
event
Gemma Model Update
-
I've shared a few deadlift demos in the past. This new demo includes some new models, accessed through the VLM Run Gateway: - SAM 3.1 to segment the barbell weight plate - ViTPose+ Large to measure the hip hinge angle, which serves as a ba…
-
640tok/s on Gemma 4 with a megakernel (withhopper.com via hn) vllmgemmamoe 6w
A worklog on megakernels for Gemma 4 on TPU v5e: 640 tok/s per user on the 26B-A4B MoE model versus 173 on an H100 with vLLM, why the dense 31B model gains only 1.2x, and when a megakernel is worth building.
-
Gevva0 – a Jev like decision engine on Gemma 26B via direct logit scoring (github.com via hn) gemmallama 6w
Gevva0: Calibrated Local Decision Gateway -00f2fe) []() A high-speed, local decision gateway running on Gemma 4 (llama.cpp / CUDA) that replaces flaky autoregressive JSON generation with direct token logit scoring, mathematical cyclic debi…
-
When a language model receives two conflicting documents as input, how does it decide which one to prioritize? Does it rely on how the sources are framed or the presentation order of the documents?